Introduction

What is GrADyS-SIM NextGen?

GrADyS-SIM NextGen is a network simulation framework. You can use this software to create simulations capable of representing scenarios populated by network nodes that coexist in a simulated environment and can send messages to each other. The primary focuses of the simulator are usability and flexibility.

Usability is achieved by opting for a simple-to-use python simulation API that allows you to quickly build and prototype decentralized algorithms in a network simulation environment. Flexibility is achieved by enabling the user to reuse his algorithm implementations on a more realistic network simulation scenario by integrating with OMNeT++ and, in the future, even reuse his algorithms in the real world through a MAVLINK integration.

The framework was created to be used in three main ways: prototype mode, integrated mode and experiment mode. As mentioned one of the main features of this framework is that you can use the exact same implemented logic on all three modes, without changing a line of code. You can understand more about these modes in the execution mode documentation here.

What's being executed in these three modes is a central piece of code called protocols. Protocols are user-defined classes that implement some kind of logic, they dictate the behaviour of a simulation node. A simulation node is an entity that exists within the simulation environment. It interacts with the environment through movement and with other nodes through network communication.

The way protocols are build they can interact with the environment without being dependent on a specific environment implementation. They observe and act through interfaces provided to them during execution in a technique commonly called dependency injection. What this means is that you can re-utilize that same code in completely different environments as long as someone has done the work of integrating that environment with the interfaces that the protocol expects. GrADyS-SIM NextGen provides integrations to three environment types which are the previously mentioned modes.

History

GrADyS is a project member of a laboratory called LAC from PUC-Rio university. Our main work is focused on exploring the coordination of autonomous vehicles through descentralized algorithms.

We created GrADyS-SIM to enable us to test our algorithms in a controlled, cheap and fast environment. Field tests are expensive and consume large amounts of time. Having a simulation environment helps us understand and validate things before we take them to the field.

GrADyS-SIM was created first and foremost to serve as a tool for the GrADyS project. As we developed it we noticed that this could be a very useful tool not only for us but for the scientific community in general.

GrADyS-SIM runs in the OMNeT++ which is an event-based network simulator that enabled us to quickly model very complex network architectures and test the effects ofdifferent scenarios, network protocols and components on our algorithms. OMNeT++ although stillvery useful has a very steep learning curve, even setting it up is a very involved process thatrequires specific setups to get it working perfectly. Even for the people inside our project it is cumbersome to work with so we concluded that some new tool needed to be built to remedy this.

These were our motivations to create this tool. We are also taking extra care to make sure this won't be a tool only for the GrADyS project, but for anyone with similar interests to ours. This means creating a very general implementation that can be used to implement all kinds of different ideas, instead of something focused only on our area of research.

GrADyS-SIM-NextGen was envisoned as a low-dependency, light and easy to use simulator that would not substitute OMNeT++ but fill the niche of being a tool for quick prototyping before more realistic results would be collected from OMNeT++. An issue we already had was translating results from the OMNeT++ simulation to the real world. Introducing a new link in this chain would only make this disconnect between our imlpementation environments greather, so GrADyS-SIM NextGen would need to somehow generate code that could be run everywhere, in python, OMNeT++ and the real world.

What should I expect from GrADyS-SIM NextGen?

GrADyS-SIM NextGen is a tool for implementing and simulating decentralized algorithms. It shines in simulating environment populated by nodes capable of communication and movement. In this dynamic environment you can implement algorithms that coordinate the movement of these nodes and the communication between them in order to achieve some goal.

The simulator will help you to implement your algorithm's logic, testing it, validating it and understanding it. You will be able to see how your algorithm performs in situations analogous to the ones it would find in the real-world. This process will help you create more resilient, robust and efficient algorithms that perform well within the constraints of the real world.

GrADyS-SIM NextGen is a low-dependency, light and easy to use simulator. It was built with simplicity of use as a primary goal. If you want quick prototyping and testing of your algorithms, this is the tool for you. We also provide more complex tools that can be used to simulate more realistic scenarios, but they are not as easy to use as the prototype-mode simulator. The algorithm logic you implement in this simulator, which we tend to call "protocols", can be reused in any of our provided simulation environments and even in the real-world, although this feature is not ready yet. All of this with no code adaptation necessary as the protocols run isolated to the code that make each of the environments work.